A5043694321- EEG and Brain-Computer Interfaces
- Blind Source Separation Techniques
- Machine Learning and ELM
- Muscle activation and electromyography studies
- Gaze Tracking and Assistive Technology
- Neonatal and fetal brain pathology
- Emotion and Mood Recognition
- Advanced Memory and Neural Computing
- Face and Expression Recognition
- Air Quality Monitoring and Forecasting
- Neural Networks and Applications
- Hand Gesture Recognition Systems
- Neural dynamics and brain function
- Advanced Sensor and Energy Harvesting Materials
- Functional Brain Connectivity Studies
- Heart Rate Variability and Autonomic Control
- Radiomics and Machine Learning in Medical Imaging
- Anomaly Detection Techniques and Applications
- Air Quality and Health Impacts
- Lung Cancer Diagnosis and Treatment
- Domain Adaptation and Few-Shot Learning
- Vehicle emissions and performance
- Brain Tumor Detection and Classification
- Neuroscience and Neural Engineering
- Sleep and Work-Related Fatigue
Hangzhou Dianzi University
2016-2025
Wenzhou Central Hospital
2025
Zhejiang Lab
2023-2024
Zhejiang University
2022
Guidewire (United States)
2022
Cytoskeleton (United States)
2022
Motor Imagery (MI) paradigm is critical in neural rehabilitation and gaming. Advances brain-computer interface (BCI) technology have facilitated the detection of MI from electroencephalogram (EEG). Previous studies proposed various EEG-based classification algorithms to identify MI, however, performance prior models was limited due cross-subject heterogeneity EEG data shortage for training. Therefore, inspired by generative adversarial network (GAN), this study aims propose an improved...
Support vector machines are powerful tools used to solve the small sample and nonlinear classification problems, but their ultimate performance depends heavily upon selection of appropriate kernel penalty parameters. In this study, we propose using a particle swarm optimization algorithm optimize both parameters in order improve support machines. The optimized classifier was evaluated with motor imagery EEG signals terms prediction. Results show that can significantly accuracy signals.
The rapid development of the automotive industry has brought great convenience to our life, which also leads a dramatic increase in amount traffic accidents. A large proportion accidents were caused by driving fatigue. EEG is considered as direct, effective, and promising modality detect In this study, we presented novel feature extraction strategy based on deep learning model achieve high classification accuracy efficiency using for fatigue detection. signals recorded from six healthy...
Emotion recognition is important in the application of brain-computer interface (BCI). Building a robust emotion model across subjects and sessions critical based BCI systems. Electroencephalogram (EEG) widely used tool to recognize different states. However, EEG has disadvantages such as small amplitude, low signal-to-noise ratio, non-stationary properties, resulting large differences subjects. To solve these problems, this paper proposes new method on multi-source associate domain...
For solving the problem of inevitable decline in accuracy cross-subject emotion recognition via Electroencephalograph (EEG) signal transfer learning due to negative data source domain, this paper offers a new method dynamically select suitable for and eliminate that may lead transfer. The which is called domain selection (CSDS) consists next three parts. 1) First, Frank-copula model established according Copula function theory study correlation between target described by Kendall...
Intermuscular coupling analysis (IMC) provides important clues for understanding human muscle motion control and serves as a valuable reference the rehabilitation assessment of stroke patients. However, higher-order interactions microscopic characteristics implied in IMC are not fully understood. This study introduced multiscale intermuscular framework based on complex networks with O-Information (Information About Organizational Structure). In addition, to introduce neural information, sEMG...
The application of hidden Markov model (HMM) to recognize gait phase using electromyographic (EMG) signals is described. Four time-domain features are extracted within a time segment each channel EMG preserve pattern structure. According the division cycle, structure HMM determined, in which state associated with phase. A modified Baum-Welch algorithm used estimate parameter HMM. And Viterbi achieves recognition by finding best sequence assign corresponding phases given segments. feature set...
The use of transfer learning in brain-computer interfaces (BCIs) has potential applications. As electroencephalogram (EEG) signals vary among different paradigms and subjects, existing EEG algorithms mainly focus on the alignment original space. They may not discover hidden details owing to low-dimensional structure EEG. To effectively data from a source target domain, multi-manifold embedding domain adaptive algorithm is proposed for BCI. First, we aligned covariance matrix Riemannian...
One major challenge in the current brain–computer interface research is accurate classification of time-varying electroencephalographic (EEG) signals. The labeled EEG samples are usually scarce, while unlabeled available large quantities and easy to collect real applications. Semi-supervised learning (SSL) methods can utilize both data improve performance over supervised approaches. However, it has been reported that may undermine SSL some cases. To safety SSL, we proposed a new...
Classification of motor imagery (MI) electroencephalogram (EEG) plays a vital role in brain-computer interface (BCI) systems. Recent research has shown that nonlinear classification algorithms perform better than their linear counterparts, but most them cannot extract sufficient significant information which leads to less efficient classification. In this paper, we propose novel approach called FDDL-ELM, combines the discriminative power extreme learning machine (ELM) with reconstruction...